OpenAI
Chain-of-Thought Reasoning: Advances and Hidden Flaws
New research published in May 2026 shows that chain-of-thought reasoning in models like DeepSeek-R1 amplifies position…
New research published in May 2026 shows that chain-of-thought reasoning in models like DeepSeek-R1 amplifies position…
Zyphra's ZAYA1-8B demonstrates efficient reasoning with only 760M active parameters while matching larger models. New research…
New research reveals that AI reasoning models exhibit increased position bias as their thinking processes grow…
New research reveals reasoning AI models face significant challenges including creative problem-solving limitations, increased position bias…
New research reveals that longer reasoning trajectories in AI models correlate with increased position bias, challenging…